When I was young and naive I voted to remain in the EU in the brexit referendum. Since then I’ve learned a few things and realise the EU is not an institution to be affiliated with politically. In terms of what we did with the opportunity we had - we totally dropped the ball.
It’s just so disappointing how the EU is still so popular across Europe.
The only way to solve this is to have the user of the credit/debit card pay the fee. Sure, you can do an EU thing of 0.2-0.3% or whatever it is, but this might still be 0.2-0.3% more than it could be in a competitive market.
Australia tried this, but it led to a different problem of payment processors being able to gouge card payers because merchants were allowed to pass on whatever their costs were, and consumers didn't really have a real choice of paying the fee or not. Also since the fee had to be stated before the card was presented for tap and go payments, it didn't solve the OP problem: debit cards and premium cards all got hit with the same surcharge.
Interesting. I guess one would then ask why was it not the case that the merchants were not also behaving in a similar way and looking for the cheapest payment processors (as long as they were also asked to eat their own side of the fees).
Perhaps forcing everyone to just implement free payments between any bank accounts, like pix, is actually the only viable solution here (and i just accept that this is one of those market failure things).
Not publicly acknowledging how misallocation of capital may be happening today shows he is corrupt. He’s not that dumb to not know it’s a major risk to the whole story, and is certainly financially incentivised to write as he does.
With the corollary that old hardware valuations will plummet with them.
Although given we have marginal pricing we need to push through to those lower prices in the face of increasing demand, so timing of this is uncertain and the key to the AI financial markets
I think what is going on here is that OpenAI are employing price differentiation to capture more of the market. The captive API customers are already there. There are bunch of potential customers that are price sensitive and are at openrouter. This move allows them to capture more of the market and increase profits (yes, they might be increasing profits at this level)
One of the selling points from openrouter is exactly that though, it makes it extremely easy to jump from model to model. For some background tasks for example I setup all the free models on openrouter and if one is not available it jumps to the next one.
Perhaps not most but surely some, and in the future perhaps increasingly easily.
I’d have thought that even today people would validate a number of models for certain tasks and on a daily basis go for the cheapest provider when they run that task.
Here are the numbers for the US, as a percentage of GDP:
- Government Debt: 123.0%
- Tax Receipts: 17.2%
- Spending: 23.1%
- Deficit: 5.9%
- Interest on Debt: 4.2%
So yeah, 1/4 of taxes go to paying interest. To allay the debt concern crowd a bit: gdp numbers are real numbers, so inflation of 3% and growth of 1% = 4% nominal, so that deficit number actually means that next years govt debt as percent of gdp won’t be materially higher.
This is the government playbook: create actual inflation of 6% per year, with reported statistic inflation of 3% per year. This means real growth looks like +3% before you need to talk about contractions/recession.
All this means that the sovereign crisis is not near and the government steals your savings at 6% per year.
Is the only reason that a behind the meter subsidy is more efficient because it reduces the upfront for an equivalent kWh storage (with the consumer footing the difference)? I would have thought few, larger, batteries at local substations was more efficient otherwise.
Distribution infrastructure capacity is a current and ongoing concern for summer peak loads (every house turning on air conditioners at once) even with the large government subsidies that have existed for residential solar panels, residential batteries, reverse cycle heat pump cooling/heating and insulation. The cost to rebuild every distribution substation, change every pole-top transformer and rewire suburbs would be much higher if these subsidies never existed.
For most of Australia's population you can look up an annual report that each transmission and distribution operator is required to publish annually for their predicted demand on substations, transmission lines and distribution lines. For example in Victoria this report is the TCPR, an example available at [1]. For many distribution substations, the 10th percentile maximum demand is predicted to occur in the early 2030's, meaning that at least transformers at substations would need upgrading, or the substation rebuilt or expanded. The reasons include higher density development in urban areas, electric vehicle charging and for winter, electrification of gas heating.
As an example of the impact of residential solar generation, [2] states that for an urban growth corridor of Melbourne being built largely in the era of residential solar panels being a standard housing feature, a distribution substation ATS West has 2x 150MVA transformers operating at combined peak of 200MVA, also with 125MW of embedded residential solar generation. Without that 125MW embedded residential solar generation, this substation and feeders perhaps would have had to be replaced years earlier to allow import of ~325MVA rather than ~200MVA. Also not mentioned explicitly in this report is residential battery storage which may be reducing maximum demand from predictions made before residential batteries were widely installed.
As a side note, perhaps Australia perhaps has the most open and transparent electricity market in the world? I don't know any other country where the entire electricity grid down to 230/400VAC distribution lines running on poles in every residential street is available publicly on a map, including serial numbers of poles and ratings of pole-top transformers. And substation schematics and constraint reports, etc are all available through various public reports. I think the transparency demonstrates a great deal of confidence in the electricity grid and how it is managed.
I've contributed a bit on the infrastructure side of things to OpenStreetMap--high voltage transmission lines, pipelines, etc. I also have contributed to alltheplaces.xyz which whilst mostly known for collating shop locations, also collates infrastructure point data such as substations, traffic cameras, bins in parks, etc. As I'm in Australia, I naturally have more of an interest in Australia.
Sometimes when trying to improve OSM with a new substation, wind farm or otherwise, you might have missing information such as a wind farm that is covered with stitched imagery--left half is a 2024 swath showing construction underway and concrete foundations prepared, or a pad excavated, right half is a 2026 swath showing turbines installed and operating. Sometimes it can be helpful to double check assumptions made about an excavated pad being a wind turbine foundation against public data--e.g. press release stating how many turbines have been ordered for the wind farm, or a low resolution map in a company report indicating rough placement of turbines.
Or for a new substation with a 330kV bus and 110kV bus confirmed by tracing transmission lines to other substation buses, there could be a dangling transmission line traced to a tower next to the new substation, but it's not clear which bus it may connect to. This is where some more detailed public data, in the rare case it may be available (such as in Australia), can be useful to combine together to confirm an assumption. A real case I found recently was in Saudi Arabia where a press release from a company stated something to the effect of "We just successfully installed a 330kV feeder 9281 to substation G40", which confirmed how the the transmission lines connected to the substation.
I have however mostly mapped in countries where OSM is most lacking--and for electricity infrastructure that is China--by orders of magnitude. An example I worked on previously is Hainan, which was previously mostly missing from OpenInfraMap.[2] China publish nothing and even require their street view imagery providers to blur substations. Even so--this is largely a pointless effort because with a bit of extra puzzle solving (such as switching between 4 historic satellite images to find the one where a shadow or a tower or pole is visible) it's almost always possible to get an accurate map produced. Gaps are mostly limited to out of date satellite imagery which for China is a huge accuracy problem because the scale of electricity grid expansion is absurd. It's commonplace to start tracing a transmission line to only find half way along, the transmission line disappears on a stitch in satellite imagery, and then more recent imagery on the other side of the stitch shows 3 new 500kV+ (including HVDC) transmission lines routed in parallel, replacing the old line. You can open an OSM editor almost anywhere in China and scroll around for a minute to find an unmapped high voltage transmission line, wind farm, substation, etc (having a list of coordinates where new satellite swaths are available would help most here).
For a global list of electricity grid data sources, [1] is very detailed list that is well maintained. Countries in Africa are particularly challenging--some you'd be lucky to find even a single 800x800px image of the electricity grid... or even a website for the transmission network operator.
Does anyone know how the grok allowances compare to OpenAI / Anthropic for the monthly plans? I heard they're not generous, which means I never really bother testing Grok.
It’s just so disappointing how the EU is still so popular across Europe.